Note: Descriptions are shown in the official language in which they were submitted.
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
REAL-TIME METHOD FOR MAINTAINING FORMATION STABILITY
to
Related Application
This application claims priority to U.S. Provisional Patent Application No.
60/209,488, filed June 6, 2000.
Background of the Invention
1. Field of the Invention
This invention relates to methods for predicting and maintaining wellbore
stability during
drilling, well servicing and production in subterranean formations,
particularly hydrocarbon
bearing subterranean formations.
2. Description of Relevant Art
As used herein, the term "(in)stabilit~' shall be understood to mean
"stability or
instability." Wellbore (in)stability in shales is a major problem costing the
petroleum industry,
according to conservative estimates, $700 million annually. Understanding and
modeling
mechanisms of shale (in)stability is an ongoing industry effort. Drilling a
hole into a formation in
2o equilibrium induces stress concentration in the vicinity of the borehole.
Interactions will occur if
parameters such as chemical potential, electrical potential, thermal
potential, ionic concentration,
etc., of the drilling fluid and the shale formation fluid are not in
equilibrium. Any differences in
these parameters will alter the near wellbore pore pressure which in turn will
influence the
borehole equivalent stress-state and the shale mechanical strength thus
affecting wellbore
(in)stability. Tests conducted at compressive uniaxial and triaxial stress
conditions in the past have
indicated the strong influence of these parameters on shale strength.
Wellbore instability predominantly occurs in chemically reactive shales that
overlay
reservoirs. A major concern of the drilling engineer is either keeping the
borehole wall from
collapsing (packing off) or fracturing (losing circulation).
3o Past efforts to develop improved well bore models ("WBlVp') for shale
drilling have
been hampered by a limited understanding of the drilling fluid/shale
interaction phenomenon.
This limited understanding has resulted in drilling fluids being designed with
properties
inadequately or insufficiently optimized to prevent the onset of borehole
(in)stability
problems. Historically, wellbore (in)stability problems have been approached
on a trial-and-
error basis, going through a costly mufti-well ,learning curve before arriving
at reasonable
I
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
solutions for optimized operations and systems.
Recent studies of fluid-shale interactions have produced fresh insights into
the
underlying causes of borehole (in)stability; and chemical potential related
instabilities in
shales have been identified. A chemical potential borehole stability model,
derived for an
arbitrary borehole orientation, has been successfully implemented to address
borehole
(in)stabilities. This model, based on a well-established elastic modeling
approach, is
discussed in Tare, U.A., and Mody, F.K.: "Novel Approach to Borehole Stability
Modeling
for ERD and Deepwater Drilling", paper SPE 52188, 1999 SPE Mid-Continent
Operations
Symposium, Oklahoma City, USA, March 28-31, 1999, incorporated herein by
reference.
io Advances in technology are providing wider and greater capabilities for
new, more
reliable, accurate and rugged downhole sensors and tools. Such apparatuses and
sensors are
increasingly able to collect and transmit information to the surface as it is
happening.
With increasing economic demands on reducing rig downtimes associated with
borehole
instability problems, there is increasing need for the ability to predict and
resolve instabilities in the
field using MWD/LWD (measuring while drilling/logging while drilling) data in
conjunction with
prior geophysical and shale/fluid interaction k nowledge. However, previous
attempts have
focused on earth stresses, using drilling data and borehole (in)stability
modeling on a "post-
mortem" basis, i.e., after substantial problems have occurred. There is a need
for methods that
enable instability problems to be avoided.
2o Summary of the Invention
T
A method ~s disclosed for maintaining subterranean formation stability, and
particularly
wellbore stability during drilling, well completion and servicing operations,
and even during
production and enhanced recovery operations. The method of the invention
affords maintenance
of wellbore stability throughout the life of a well penetrating a subterranean
formation.
Moreover, the method provides for real-time monitoring of fluid-formation
interaction such that
adjustments can be made before (in)stability problems occur.
In the method, initial formation characteristics or parameters are estimated
or obtained
and the rock strength of the formation is estimated. Characteristics of the
fluids injected into the
wellbore or the formation during drilling, or during well completion and
servicing operations, or
3o during production or enhanced recovery operations, are also estimated or
obtained. This data
concerning the formation and the fluids is input into a well-stability model
and results are used to
revise the estimates of the formation parameters and/or the fluid parameters
or to adjust the fluid
2~
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
composition or characteristics. These parameters and the model results are
updated with real
time formation measurements taken during the particular operation being
conducted (i.e., drilling,
well servicing, or production). Continued acquisition and processing of the
formation data in the
model allows the fluids to be adjusted as needed to avoid formation
instability problems caused by
interaction of the fluids with the formation.
Brief Descrietion of the Drawings
Figures 1A and 1B are a flowchart or process map for a real-time wellbore
stability
process according to the invention.
Figure 2 is a schematic of a Pore Pressure Transmission experiment designed to
study
1o time-dependent alterations in shale properCies as a function of drilling
fluid exposure.
Figure 3 is a graph of the time dependent transient pore pressure response of
Pierre II
shale when exposed to a 20% w/w NaCI solution.
Figure 4 is a graph of the time dependent transient pore pressure response of
Pierre II
shale when exposed to a sodium silicate solution containing 20% w/w NaCI.
15 Figure 5 is a graph of the time dependent transient pore pressure response
of Pierre II
shale when exposed to a polymer in solution containing 20% w/w CaCla salt and
2% KCl salt.
Figure 6 is a graph of the shale compressional acoustic wave response of
Pierre II shale
when exposed to a 20% w/w NaCI solution.
Figure 7 is a graph of the shale shear acoustic wave response of Pierre II
shale when
2o exposed to a 20% w/w NaCI solution.
Figure 8 is a graph of the shale compressional acoustic wave response of
Pierre II shale
when exposed to a sodium silicate solution containing 20% w/w NaCI.
Figure 9 is a graph of the shale shear acoustic wave response of Pierre II
shale when
exposed to a sodium silicate solution containing 20% w/w NaCI.
25 Figure 10 is a graph of the compressional acoustic wave response of Pierre
II shale when
exposed to a polymer in solution containing 20% w/w CaCla salt and 2% KCl
salt.
Figure 11 is a graph of the shear acoustic wave response of Pierre II shale
when exposed
to a polymer in solution containing 20% w/w CaCla salt and 2% KCl salt.
Figure 12 is a graph of dynamic Young's modulus as a function of transient
pore pressure
3o for Pierre lI shale exposed to a 20% NaCI solution.
Figure 13 is a graph of dynamic Young's modulus as a function of transient
pore pressure
for Pierre II shale exposed to a sodium silicate solution containing 20% w/w
NaCI.
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
Figure 14 is a graph of dynamic Young's modulus as a function of transient
pore pressure
for Pierre It shale exposed to a polymer in solution containing 20% w/w CaCla
salt and 2% ICI
salt.
Detailed Description of Preferred Embodiments
The following terms as used herein, and particularly as used in the figures,
have the
meaning indicated below:
Nomenclature
o =.llslembrar2e e~ciency,
P = Pressure, mlL~, psi
to E = Young's modulusna mlLtz, psi
Aw = Water activity
dP = Pressure drop, mlLtz, psi
L' = Acoustic Ijelo,~ity, mlt, nalsec
Subscripts
is = confzning
o = external of° pore fluid
fl = test fluid
sh = shale fluid
exp = experimental
2o p = sonic compressional waves
s = sonic shear waves
SI Metric Conversion Factors
ftx3.048 E-0I=m
in. ac 2. 54 E+00 = cm
lbf x 4.448 222 E+00 = N
psi x 6. X94 757 E+00 = kPa
In investigating the role of various potentials of drilling fluids on
transient rock properties
of shale and other subterranean formations subjected to compressive confining
stress, we have
3o discovered that properties of cores from such formations should be measured
as a function of
time when exposed to a drilling fluid, well service fluid, or enhanced
production fluid, or even
when subjected to production itself. Experiments were conducted under
conditions simulating a
borehole portion under geostatic stress with the wellbore surface exposed to a
circulating drilling
fluid. The results supported our discovery.
4
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
In the experiments, rock properties such as acoustic velocity and transient
pressure for
each core sample were measured simultaneously as a function of time during the
experiments.
This simultaneous measurement eliminated the anisotropy associated differences
obtained when
' different samples are used to determine the relationship between these
characteristic parameters of
the subterranean formations, and in these experiments, particularly the
characteristic parameters of
the shale formations. This testing enhanced understanding of the relationship
between acoustical,
mechanical, chemical, electrical and failure properties of the shale as a
function of time when
exposed to a fluid, and in these experiments, particularly when exposed to a
drilling fluid. Such
measurements provide the necessary precursor to determine the onset of time
dependent
1o in(in)stability, thus providing options to implement corrective measures in
real time to alleviate
(in)stability, accordixig to the method of the present invention.
Without wishing to be limited by theory, we believe, based on our observations
from
these above experiments, that measurement of rock properties such as acoustic
velocity as a
function of the transient pressure of shales under simulated in-situ stress-
state facilitates real time
wellbore (in)stability logging using MWD/LWD or any other relevant
information. This practice
provides better real-time management of (in)stability problems in the field.
The present invention provides a method or process for maintaining formation
stability
through monitoring the interaction between the formation and fluids used in
well drilling,
servicing or production. Further, the present invention provides a real-time
method for improved
2o drilling, servicing aid producing wellbores in or through subterranean
formations by employing
such monitoring and maintaining of well stability. The improvements of the
invention are
especially appreciated in operations in shale formations, although the
invention is applicable as
well to other types of subterranean formations. Similarly, the improvements of
the invention are
especially appreciated in drilling through subterranean formations, although
the invention is
applicable as well to other operations in subterranean formations, such as
well completion and
well servicing operations, and even to producing operations. Thus, the
advantages of the
invention are best xealized when the invention is applied throughout the life
of a wellbore in a
subterranean formation, and especially a hydrocarbon bearing subterranean
formation.
The method of the invention employs real-time wellbore stability modeling.
Figures 1A
3o and 1B present a process map of real-time wellbore stability modeling,
specifically for example as
applied to drilling operations. A similar map would be used for well
completion or other service
operations employing well service fluids instead of drilling. fluids. Also, a
similar map would be
5
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
used for well production operations, monitoring stability of a well as fluids
may be produced from
a formation, and monitoring stability of a well during various enhanced
recovery operations (for
enhancing recovery of hydrocarbons from the formation). In all such cases, the
interaction of the
formation to various fluids being introduced into or being produced from the
wellbore are
monitored. Hereinafter, the description of the invention will primarily be
directed to drilling
operations and to drilling fluids for example. However, the invention is not
limited to such
operations or fluids and may be similarly applied to well completion and
service operations
involving well service fluids and to production operations.
The initial step in the method or process of the invention involves
constructing an earth-
1o model (preferably including stresses, formation compressive and tensile
strengths and geologic
features) from a variety of sources such as pre-drill, seismic, and offset
data. Based on the initial
earth-model, predictions are made for mud weights, casing shoe locations,
drilling fluid type and
chemistry, etc. as would be known to those skilled in the art. Preferably
simultaneously,
laboratory tests are conducted for specific formations (e.g., shales) that are
expected to be drilled,
is using either actual bores from such formations or representative outcrop
samples. The tests are
conducted using a variety of experimental apparatuses such as the Pore
Pressure Transmission
(PPT) test apparatus. Information on time-dependent formation (e.g., shale)
behavior is obtained
from such tests.
Further in the method of this invention, rock properties such as transient
pore pressure,
2o compressional and shear wave velocities, deformation, and strength of shale
cores under triaxial
stress are measured. These measurements are used to calibrate the wellbore
stability model (also
called the total potential model), such as the model shown in the flow
diagrams in Figure 1B, for
specific application, such as, for example, drilling with programmed or
planned drilling fluids
throughlinto a reservoir overlaying shales or other formations of interest.
Measurements of static
25 and dynamic rock properties as a function of the transient pressure of
shales (or other formations
of interest) under triaxial stress state enhances understanding of the
relationship between these
parameters. Such measurements also help in determining how economically to
obtain these
properties using MWD/LWD (measurement while drilling/logging while drilling)
data.
During the drilling process, the earth-model and the total potential wellbore
stability
3o model are updated with real-time measurements while dmlling (MWD) andlor
using wireline logs
or any other techniques for determining static and dynamic rock properties.
The captured MWD
and/or wireline date. or data obtained from any other techniques is processed
in real-time and this
6
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
data is converted into input parameters for the wellbore stability model. The
input parameters are
thus calibrated with MWD and/or wireline data or data obtained from other
techniques. The
calibrated input parameters are then used in a wellbore stability model along
with the laboratory
time-dependent measurements.
For a given set of input parameters the wellbore stability model provided in
Figure
1B can perform in three modes of analysis. These are briefly described below:
Safe mud weight window prediction - Results obtained from this mode provide
the
upper and lower limit for a safe mud weight for a given composition of the
drilling fluid
as a function of the hole inclination angle and orientation. The upper limit
in this case is
to the mud weight above which extension fracturing or fracture propagation
would occur
and would result in excessive drilling fluid losses. The lower limit is set by
formation pore
pressure or the minimum mud weight required to prevent borehole collapse,
whichever is
greater. In this mode, chemical and thermal interaction between the formation
(mostly
shale) and the drilling fluid can be incorporated.
is . Optimum chemical composition prediction (OBM) - In this mode of analysis
the model
provides the optimum chemical composition of the drilling fluid for a given
mud weight,
hole inclination and direction. In the absence of laboratory derived shale
pore fluid
potential, in-situ shale pore fluid potential can be predicted using a sub-
module.
Optimum chemical composition of drilling fluid prediction (OB1V1) followed by
safe mud
2o weight window prediction - . The model in this mode first optimizes the
chemical
composition o~ the drilling fluid for the actual (i.e., utilized currently in
the field) mud
weight followed by a prediction of the safe mud weight window as a function of
hole
angle for a given direction.
All three modes have access to an empirical rock strength estimation module,
as shown in
25 Figure l A.
The predictions from the wellbore stability model are sent to a visualization
module
(typically graphing of results) based on one of two types of activities-field
implementation or
feedback, as shown in Figure lA-to be executed. These activities provide
feedback into the data
collection, processing and interpretation process and field implementation of
the wellbore stability
3o model results, as shpwn in figures 1A and 1B.
Wellbore Stability Model
The rationale for the wellbore stability model (total potential model) is as
follows:
7
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
The fundamental direct and coupled flow phenomena, induced hydraulic
gradients, chemical
potential, electrical potential gradients and temperature gradients all
influence rack properties and
wellbore or formation stability. In most practical cases, however, the flow
phenomena induced by
the hydraulic gradient (i.e., mud filtrate invasion driven by overbalanced mud
weight) and the
chemical potential difference between the drilling fluid and shale (or
formation) pore fluid (e.g.,
osmotic or diffusion transport of water molecules and/or ions and/or
(hydrated) solutes in and out
of the shale, electrical potential, thermal potential, etc.) are generally
considered to be the more
relevant driving forces for formation (in)stability. At any given time during
the drilling operation,
the (in)stability issue is ultimately controlled by the relationship between
the borehole stress-state
to and the rock strength. As the rock strength is inversely related to the
water content of the shale
(or of the formation of interest), modification of pore pressure is a
fundamental parameter in
altering the effective stress-state around the wellbore. Adequate mud weight
density is also
needed for borehol~ stability.
The model used in the method of this invention is preferably based on well-
established
modeling principles such as described in Mody, F.I~. and Hale, A.H.: "A
Borehole Stability
Model To Couple the Mechanics and Chemistry of Drilling Fluid/Shale
Interaction", paper SPE
25728, 1993 IADC/SPE Drilling Conf., Amsterdam, Feb. 23-25, 1993; Bradley,
W.B.: 'failure
of Inclined Boreholes' ; J. Energy Res. Tech. {Dec. 1979) 232: Trans.,
AZ1VV1E, 101; Tare, U.A.,
and Mody, F.K.: '4lVovel Approach to Borehole Stability Modeling for ERD and
Deepwater
2o Drilling", paper SPF 52188, 1999 SPE lVfid-Continent Operations Symposium,
Oklahoma City,
USA, March 2831, 1999; ~' Tan, C.P., et al.: "Integrated Rock Mechanics and
Drilling Fluid
Design Approach ~o Manage Shale Instability", paper SPE 47259, SPE/ISRM Eurock
'98,
Trondheim, Norway, Jul. 8-10, 1998; and Ghassemi, A., et al.: "A Solution for
Stress Distribution
Around An Inclined Borehole in Shale", paper No. 043, Int. J. Rock Mech. Min.
Sci. Vol. 35,
No. 415, p.538-540, 1998, which are all incorporated herein by reference.
The first two of these references listed above have pertinent theoretically
derived
solutions to the model. The model incorporates chemical potential effects
using the
difference in the molar free energies (i.e., activity) of the drilling fluid
and the shale pore
fluid. Also incorporated in the model is the concept of reflection
coefficient, 6, as discussed
3o in Staverman, A.J.: "Theory of Measurement of Osmotic Pressure", Recueil
des Travaux
Chimiques des Pays-Bas, v.70, pp 344-352, (1951), incorporated herein by
reference. The
temperature difference between the formation and the drilling fluid directly
affects the near-
8
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
wellbore stress concentration. The model utilizes this temperature difference
and calculates
the effect of thermal stresses on the near-wellbore tangential stress where
significant.
Egneriments
Pore Pressure Transmission (PPT) Tests
The Pore Pressure Transmission (PPT) experiment has been designed to study
time-
dependent alterations in shale properties as a function of drilling fluid
exposure. A schematic of
this test is shown in Figure 2. The experiment consists of confining well-
preserved shale cores or
outcrops, under geostatic stress and then circulating (under confined dynamic
conditions) test
to fluids at the upstream side while measuring changes in the downstream
pressure. The upstream
pressure may be increased to simulate overbalance conditions representing
downhole conditions.
Typically during a drilling operation, hydraulic overbalance is maintained to
provide the formation
with net radial support. The downstream pressure indicates changes in the
sample pore pressure
at some distance into the wellbore.
Experimental Setup for PPT Tests
The experimental setup consists of a jacketed steel pressure vessel with built
in pressure
ports, connections and acoustic transducers. The sample (1 in. x 2 in:) is
jacketed with a Teflon
membrane after six extension-meter strain gauges are applied to measure the
sample deformation.
The entire assembly is placed in a temperature regulating device to ensure
adequate control of the
2o experimental temperature.
Summary of PPT Test Procedure
A shale sample is equilibrated at a hydrostatic confining pressure P;~ (e.g.
500 psi), under
external fluid pressure po (e.g., 50 psi) on both ends, and at the specific
test temperature. Fluid
pressure is applied with simulated pore fluid. Once saturation is achieved,
the upstream pressure
line is closed. The confining pressure is then increased to the required level
for consolidation, P;
(e.g., 4000 psi). The system remains closed until the pressures at both the
sample ends have
stabilized. Upstream and downstream pressure is allowed to stabilize and
equilibrate.
Upon stabilization of the pore pressure, the sample is allowed to consolidate.
The
downstream pressure of the sample is monitored.
3o After the pressures at both ends have stabilized, the pore fluid is
replaced with a drilling
fluid that is circulated at a constant pressure. The transient pressure
behavior, sonic compression
and shear travel t:~mes, radial and axial deformation, temperature, and
circulation rate are
9
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
monitored.
PPT Test Details
The PPT tests provide a database including shale responses to various drilling
fluids and
salt solutions. Fox example, the PPT tests provide meaningfi~I membrane
efficiency numbers,
allow for the study of more complex interaction between shale and drilling
fluid to validate,
verify, or confirm existing chemical potential poroelastic models, enable
calibration of the shale
response to known drilling fluids for optimum application in the field, and
demonstrate
simultaneous data acquisition of various responses to provide detailed shale
property
characterization as a function of drilling fluid exposure.
to The three example fluids tested were a 20% w/w sodium chloride salt
solution, a sodium
silicate system corr~prising 20% w/w sodium chloride, and a polymer in
solution comprising
20% w/w CaCla salt and 2% KCI salt. It is important to establish a baseline of
the shale response
to the cation or canons that would be included in the drilling fluid system.
Such a baseline helps
in identifying or isolating shale alterations due to cation exchange. In
addition this baseline helps
differentiate, to a certain extent, the effects of the base salt solution and
the effects of other shale
modifying additives (e.g., silicates, polymers like partially- hydrolyzed poly-
acrylamide,
amphoterics etc.).
Discussion of Results
Based on the transient pore pressure response of Pierre II shale exposed to
the test fluids,
2o the sodium silicate solution comprising 20% w/w NaCI showed higher membrane
efficiency
numbers than the 20% w/w NaCI solution and the polymer in solution comprising
20% w/w
CaCla salt and 2% KCl salt (approximately 60% e$ciency versus approximately
10% efficiency
and approximately 5.30% efficiency respectively).
Figures 3, 4 and 5 show the time dependent transient pore pressure response of
Pierre IC
shale when exposed to a 20% w/w NaCI solution, a sodium silicate system
comprising 20% w/w
NaCI and a polymer in solution comprising 20% w/w CaCla salt and 2% KCl salt,
respectively. The saturating fluid (simulated pore fluid) in alI cases is 8%
w/w NaCI solution. The
shale acoustic (compressional and shear wave) response during the entire
experiment for the 20%
w/w NaCI solution is shown in Figures. 6 and 7. Figures. 8 and 9 provide the
corresponding
3o acoustic (compress~onal and shear wave) response of Pierre II shale when
exposed to the sodium
silicate solution cpmprising 20% w/w NaCI. Figures 10 and 11 provide the
acoustic
(compressional and shear wave) response of Pierre II shale when exposed to the
polymer in
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
solution comprising 20% w/w CaCla salt and 2% KCI salt. One example rock
property, the
dynamic Young's modulus (E) (reflecting rock elasticity), calculated from the
above acoustic data
with respect to the three solutions is shown respectively in Figures. 12, 13
and 14. Each of the
Figures 3-14 indicates the three stages of the PPT test. First, the shale is
saturated with 8% w/w
NaCI (simulated pore fluid); second, 8% w/w NaCI is circulated until
equilibrium; and third, the
shale is exposed to the test fluid.
Table 1 provides transient pore pressure data and corresponding calculated
membrane
effciency in tabular form for the test fluids when exposed to Pierre II shale.
TABLE 1-MEMBRANE
EFFICIENCY CALCULATION
Test Fluid Awfl Aw~n dPthepsidPeXPpsio,
dPexPld
Pthe
20% w/w NaCl solution0.83 0.95 2685 270 10.05
Sodium silicate solution0.83 0.95 2685 1604 59.74
comprising 20% w/w
NaCI
Polymer solution 0.78 0.95 3920 208 5.30
comprising
20% w/w CaCla salt
and 2%
KCI salt
to where dP~e refers to the difference or change in theoretical pressure and
dPeXp refers to the
difference or change in experimental pressure.
The well known concept of "reflection coefficient" (i.e., membrane ideality)
may be
used for "leaky systems" including water based mud (WBM)/shale (or other
formation) type
systems for formation stability applications. For WBM/shale systems, membrane
efficiency is
not a clearly defined term (e.g., as an oil fiIW on shales present in invert
oil emulsion
systems). The membrane efficiency of formation-fluid systems is due to a
difference in
mobility of water and solutes (ions) in shales. Thus membrane efficiency can
be defined as:
v solute
v water 1
If only the movement of water is allowed and all the solutes are rejected, the
membrane is ideal semi-permeable (100% membrane efficiency or 6=1). However,
for shales,
the mobility of solutes is lower than that of water and the membrane is "non-
ideal" or
"leaky", 0<6<1. A non-ideal membrane does not entirely restrict the transport
of solutes.
Membrane efficiency ("reflection coeffcient") has a relationship with at least
two key
parameters of the WBM/shale system, i.e., the average 'effective' shale pore
radius and the
radius of hydrated ions present in the drilling fluid system.
11
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
A mathematical representation to describe the driving force for the movement
of
water by an osmotic mechanism and one way or means of estimating membrane
efficiency
directly is shown i~ the following equation:
~xln~Awdj ~=t(o'xL~PP)=~cr x(P-Pp ) (2)
A wsh
where Awaf is the water activity of the drilling fluid, which can be estimated
a number of
known ways, most notably, by partial vapor pressure determination, boiling
point elevation,
or directly using a hygrometer; Awsh is the water activity of the shale pore
fluid and can be
measured (for preserved shale cores) by partial vapor pressure determination;
Pp is the far-
field pore pressure; P is the near wellbore pore pressure; and 6 is the
membrane effciency
1o term, or "reflection coefficient", specific to a shale/drilling fluid
system.
Further discussion of the calculation of membrane efllciency may be found in
references Mody, F.K. and Hale, A.H.: "A Borehole Stability Model To Couple
the Mechanics
and Chemistry of Drilling Fluid/Shale Interaction", paper SPE 25728, 1993
LA.DC/SPE Drilling
Conf., Amsterdam, Feb. 23-25, 1993, and Staverman, A.J.: "Theory ofMeasurement
of Osmotic
Pressure", Recueil des Travaux Chimiques des Pays-Bas, v.70, pp 344-352,
(1951), which are
incorporated herein by reference.
While the membrane eflxciency for a 20% w/w NaCI solution when exposed to
Pierre
TI shale is 10.05%, and the membrane efficiency for a polymer in solution
comprising 20%
w/w CaCl2 salt and 2% KCl salt when exposed to Pierre II shale is 5.30%, the
membrane
2o e~ciency for sodium silicate systems is significantly higher at about 60%.
For reference, an
oil-based mud has a theoretical membrane efficiency close to 100%. Following
the response
of a sodium silicate solution with 20% w/w NaCI, the shale pore pressure
decreases as a
function of exposure time to the solution. Comparing the results among the
three test fluids
indicates that, when membrane efficiency numbers are significantly higher,
drilling fluid water
activity that is lovyer than the shale pore fluid activity tends to reduce
pore pressure and
thereby contributes towards reducing near-wellbore effective stresses on a
time-dependent
basis.
Acquisition of acoustic data simultaneously with exposure of the rock to
drilling fluid, in
the field and in the PPT tests provides usefi~l data to calculate dynamic
changes in the rock
properties, changes that may be a result of that exposure. Comparing Figures
12, 13 and 14
12
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
shows changes in the dynamic Young's modulus as a function of transient pore
pressure. Young's
rnodulus, which is a measure of the stiffness or elastic properties of a rock,
is but one example of
a rock property. Similar computations can be made for dynamic Poisson's ratio
and other rock
properties derived from sonic data. In the example, when shale is exposed to
20% w/w NaCI
solution, the effective stress changes from 2105 psi immediately after
displacement to 2375 psi
when the transient pore pressure has reached its lowest point (Figure 3). For
the case when Pierre
II shale is exposed to the sodium silicate fluid the effective stress
increases from 860 psi
immediately after displacing the simulated pore fluid with the test fluid to
2464 psi at the lowest
point on the transient pore pressure curve (Figure 4). Inherent changes in
effective stress can be
observed from velocity data. For example, comparing Figures 6 and 10 with 8
shows a dramatic
change in the compressional wave sonic velocity when the shale is exposed to
the two test fluids.
Similar observations can be made for the shear wave velocity profiles by
comparing Figures. 7
and 11 with 9. These comparisons provide additional means of observing
transient pore pressure
A
in the shales as a function of test fluid exposure.
This test example shows that mud systems should be calibrated, preferably with
shale
cores under in-situ conditions. PPT tests can provide meaningful membrane
efficiency numbers
which can then be used in a wellbore stability model for a realistic analysis.
Additionally, a time
dependent chemical potential related shale transient pressure can be obtained,
which can be used
to analyze exposure time related borehole deterioration observed (based on the
nature of the
curve).
Input Parameters for Real - Time Wellbore Stability Modeling
Table 2 provides a list of input parameters fox real-time borehole stability
analysis
according to the method of the invention and the possible sources where the
data needed for
the analysis can be obtained. The input parameters can be divided into five
categories. These
are:
a) Insitu Stresses.
b) Formation Data.
c) Well Path Data.
d) Drilling Fluid.
3o e) Rock~Strength.
Each of the above categories can be further sub-divided into individual
subcategories. The
possible sources for obtaining information for each input parameter
subcategory are provided
13
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
in Table 2 (under the column heading Possible Source). The Input Parameters
categories are
discussed below:
(a) In-situ Stresses - These stresses affect the mud weights used to drill a
borehole. When
the stresses are higher than the rock strength failure occurs. The principle
stresses control
both fracturing as well as the collapse mode of wellbore failure. There are
three principle
earth stresses. The in-situ stresses are the minimum horizontal stress,
maximum horizontal
stress and the overburden stress. In a normal faulting environment such as
that typically
found in the Gulf of Mexico (away from any tectonic activity), the minimum
horizontal stress
is the least principle stress, the maximum horizontal stress is the
intermediate principle stress,
to and the overburden stress in the maximum principle stress. The major source
for obtaining
magnitudes of the feast principle stress in formations is by pressure testing
in open boreholes.
Information can be obtained from tests such as a leak-off test, extended leak-
off test, mini-
frac test, etc. Direction of the intermediate principle stress is obtained by
measuring the
location of borehole ovality (break-outs); the magnitude is estimated by back-
calculating the
break-out width for a given intermediate stress and the other two principle
stresses, pore
pressure and the rack strength. Comparison is made between the estimated and
the actual
observed breakout width. The overburden stress is typically obtained from an
integrated bulk
density log, by the actual measurement of drilled cuttings, etc.
(b) Formation Data - This data can be categorized into four sub-categories:
temperature;
2o rock properties; sale or pore water salinity; and pore pressure. Formation
data is different
for different formations. The formation temperature is typically estimated by
running
temperature measurement while drilling tools, infra-red techniques, etc. Rock
properties are
obtained either by direct measurement of actual cores in the laboratory or by
empirical
correlations based on wellbore measurements. Rock properties aid in modeling
deformation
behavior of specific formations. Shale activity or pore water salinity is used
to estimate the
chemical potential of the formation. A difference in potential between the
drilling fluid and
formation induces flow phenomenon from one to the other, i.e., flow occurs
from higher
potential to the source that contains lower potential. Osmotic membrane
efficiency and the
osmotic pressure are based on shale water activity. Shale water activity can
be obtained from
log based empirical correlations as provided in Table 2 or by actually
measuring the relative
humidity of a well preserved shale core in the laboratory. Pore pressure is
critical information
for drilling a well. Fore pressure affects the minimum mud weight used to
drill and is related
14
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
to total potential of the formation. Pore pressure is typically obtained from
pressure build-up
tests like drill-stem testing and repeat formation testing, as well as from
empirical
correlations using log and seismic information.
(c) Well Path - Drilling data includes well trajectory information like hole
angle, well
azimuth and other associated information on geology like presence of faults
and other
structures. Case histories of successful mud programs will also be a part of
the drilling data.
(d) lDrilling Fluid - This category includes three sub-categories. The
drilling fluid density
(Mud weight) can be captured in real-time (while drilling) by means of a
pressure while
drilling (PWD) Coal. Accurate information of mud weight is necessary to ensure
wellbore
1o stability. Field and laboratory densitometers may since if downhole tools
measuring mud
weight are not installed. Drilling fluid activity is a measure of the chemical
potential of the
drilling fluid. A difference in the activity of the drilling fluid and the
shale pore fluid would
induce osmotic fluid flow in the presence of a semi-permeable membrane.
Drilling fluid
activity can be measured directly using a hygrometer or a relative humidity
meter or
indirectly by for example estimating solute concentrations in titration
experiments.
Membrane efficiency is a critical parameter that determines the effectiveness
of a semi-
permeable membrane to allow only the transport of fluids relative to the
transport of solutes.
Membrane efficiency of a drilling fluid - shale formation system can be
estimated by
laboratory measurements in a pore pressure transmissions test (PPT) apparatus.
Time lapsed
2o measurements of shale properties of shales exposed to drilling fluids in a
borehole can also
lead to membrane efficiency estimations. Pore radii along with solute sizes
and concentration
may provide alternate ways of estimating membrane efficiency.
(e) Rock Strength - Rock strength can be differentiated between two types--the
compressive strength and the tensile strength. Rock strength is a critical
input parameter
along with the stresses and can be defined by various failure models. The
existing real-time
wellbore stability process includes but is not be limited to failure models
such as the Mohr-
Coulomb Failure Criterion, Drucker-Prager model, Modified Lade Criterion, Hoek
and
Brown Expression, Johnston and Chiu Model, Peak Strength Criterion, Uniaxial
compressive Strength limitation, etc. Ideally direct measurements of stress
versus strain on
3o well-preserved core samples in the laboratory in a poly-axial cell is
preferred. In,the absence
of core samples various tests on drill cuttings and empirical correlation's
(using ,Iag data can
provide rock strength estimations. For example if the water content of x~~
shale can be
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
estimated then a proprietary empirical correlation that uses water content and
shale
composition can be used to estimate rock strength.
Therefore,
Rock Strength oc f(Shale Property)
For example, Shale Property = Water content or Brinell Hardness or Cation
Exchange
Capacity or Porosity etc.
The numbers in Table 2 below have the following meanings:
1. X-ray Analysis of cuttings.
2. Nuclear Magnetic Resonance (NMR) logs (Wireline and/or MWD).
3. Resistivity/Conductivity logs (Wireline and/or MWD)
4. Mud Logs
S. Acoustic/Sonic compressional and shear wave logs (Wixeline and/or MWD)
6. Pressure and Sat~pling tools like drill stem testing (Wireline and/or MWD)
and repeat formation testing
(RFT) tools.
7. MDT
8. Spontaneous Polarization (SP) Curve logs
9. Seismic processing/Vertical Seismic Profiling (VSP) techniques.
10. Pore pressure ahead of the bit techniques.
11. Pressure While Drilling (PWD) logs (MWD).
12. Empirical and analytical models using drilling fluid rheology, pressure,
cuttings accumulation and
temperature.
13. Leak-off test (LOT), Extended Leak-off Test (XLOT), Minifrac tests,
Microfrac Tests, Formation
Integrity Tests (FIT), Casing Integrity Tests (GIT).
14. Anelastic Strain Recovery (ASR) Techniques (including tests performed on
cores and downhole)
2S 15. Regional bulk density trends.
16. Integrated bulls density logs (Wireline and/or MWD)
17. Acoustic, four anti six arm Caliper logs (Wireline and/or MWD),
information on breakout angles,
breakout volume end breakout length, etc.
18. Formation Evaluation (FE) Logs (Wireline and/or MWD)
19. Temperature logs (Wireline and/or MWD)
20. Neutron source (Nuclear) or Gamma-ray.logs like neutron-slow-neutron logs,
Double- or Multiple -
Spaced Neutron (~VIS Neutron Curve) - Dual-Spacing Neutxon Curve, Flowing
Neutron Log (Nuclear
Flolog - Lane-Wells), etc. (Wireline and/or MWD)
21. Formation Imaging logs (Wireline and/or MWD).
16
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
Table 2, List of input parameters for real-time wellbore stability analysis
and possible sources
where the data required to do the analysis can be obtained:
Possible Source Intermediate ParametersIn ut Parameter
13, 14, empirical and analytical Minimum
methods using log
and regional information like the Horizontal N
Daine's method,
Eaton's technique, Pennebaker method, Stress S
etc.
15, 16, empirical methods using Overburden
log information and T
compaction trends, measurements Stress
on drill-cuttings
(cuttin s densit measurements),
etc.,
14, 17, 21, fault stability and Inequality (magnitude)Maximum S
reactivation models, and
direction of horizontalHorizontal T
stresses Stress
E
S
S
E
S
19, field measurements like using Formation temperatureTemperature
infra-red and
techni ues flow-line tem erature
Laboratory analysis of cores, measurementsPoisson's Ratio,
on Young's
cuttings, pressure pulsing the well,Modulus, Thermal
2, 3, 5, 9, 13, 14,
20 Expansion coe~cient,
fluid
O
and grain volume Rock g
expansion
factor, rock and properties M
fluid
compressibility,
rock and
fluid bulls modulus, T
Skernpton's Pore
Pressure
Coefficient, Biots p
Constant,
etc.,
N
Log based empirical correlation's
using the following
measurements - 2, 3, 6, and 20.
Laboratory
measurements of shale cores using Shoe
testing methods
like adso tion isotherm method h T
rometer readin ActivitylPore
fl' ~ yg g
and other relative humidity measurements, W alter
determination of shale pore water Salinity
salinity from sand
pore water salinity overlaying and
underlying the
shales
9, 3,5, Pressure Build-up tests Pore Pressure Gradient
like 6 and 7, 4, 2, 10,
Ratio Methods, empirical methods Pore Pressure
using log
information and com action trends,
etc.,
Loggers or Drillers Depth, ReservoirDepth, Well trajectory W
modeling and
rote retation tools E
Seismic information, drilling data,_
18, etc. Geologic information
(presence of faultsDrilling
and types Data
of faults)
Case history for a local well whereSuccessful mud
no or very few programs
problems were seen
H
17
CA 02412337 2002-12-04
WO 01/94749 PCT/USO1/18134
11, 4, 12 Static and equivalentMud Weight
circulatin densit R
.
Laboratory titration experiments Drilling I
like the Baroid Fluid
method and the API method, drilling Activity
fluid daily
re orts L
Laboratory testing of preserved Time dependent changes
shale cores using in
N
experiments like the Pore Pressureformation water
Transmission content,
(PPT) test, Linear Swell Meter porosity, resistavity,Membrane
Tests, measuring sonic
changes in formation properties velocity and travelEff ciency
using field time,
information like, 2, 3, 5, 9, 13, formation densit
1b, 17, 18, 19, 20 , etc.
Microscopic Imaging Techniques, Pore radii and distribution, U
Physical Intrusion
Techniques like mercury injection,solute/water relative I
and nuclear
techniques such as using nuclear mobility
magnetic resonance
techniques (laboratory or wireline
or MWD), 2
1, 2, 3, 4, 5, 8, 9, 18 Clay Content and
Formation
T a
3, 4, Di-adsorption methods such Cation Exchange
as Methylene Blue Capacity
Tests or Compulsive Exchange Methods(CEC)/Surface Area,
like the Ionic
barium chloride compulsive exchangeStrength ModificationROCK
method
Exchan a E uations STRENGTH
2, 3, 4, 5, 8, 10, 15, 16, 18, Formation Water
20
ContentlPorosity/Penneabili
t
Laboratory analysis of cores, empirical
correlation's
using the following sources -1,
2, 3, 4, 5, 17, 18, 20
and formation dielectric constant
measurement.
In applying the method of this invention, at least some understanding of the
drilling
fluid/formation interaction should preferably be obtained prior to applying
any total potential or
wellbore (in)stability model for a real-time drilling operation. A Pore
Pressure Transmission (PPT)
test or any other relevant test can be used to estimate the impact of a
drilling fluid on the
formation propertiES i.e., including but not limited to transient pore
pressure, water content,
electrical properties, acoustic properties, etc.
Changes in the shale stress due to alterations of the above mentioned
properties can be
obtained from the simultaneous acquisition of sonic compression and shear wave
velocity data. A
time-dependent total potential related shale transient pressure obtained from
the PPT can be used
to address issues where exposure time related borehole deterioration is
observed.
The foregoing description of the invention is intended to be a description of
preferred
embodiments. Various changes in the details of the described method, including
without
limitation details concerning the wellbore stability model, can be made
without departing
from the intended scope of this invention as defined by the appended claims.
18
